EDA - Time - Clusters¶
In [172]:
# -*- coding: utf-8 -*-
"""
Analisis de la producción cientifica a traves del tiempo
"""
Out[172]:
In [173]:
%%HTML
<script src="require.js"></script>
In [174]:
# Importar librerías
import numpy as np
import pandas as pd
import matplotlib.pyplot as plt
from matplotlib.pyplot import figure
import plotly.express as px
import seaborn as sns
import plotly.graph_objects as go
import plotly.io as pio
from sklearn.linear_model import LinearRegression
pio.renderers.default='notebook'
In [175]:
from google.colab import drive
drive.mount('/content/drive')
Data¶
Cluster.infoInstituciones.csv https://drive.google.com/drive/u/0/folders/1APVnfD1zPgTHFAr0YMG6UMTwZ52MQVa9
Cluster.cluster.csv https://drive.google.com/drive/u/0/folders/1APVnfD1zPgTHFAr0YMG6UMTwZ52MQVa9
In [176]:
from google.colab import drive
drive.mount('/content/drive/')
In [177]:
%cd /content/drive/My Drive/3BIO-Cluster-DataAnalysis
In [178]:
%ls
In [179]:
#Data
df = pd.read_csv("Cluster.infoInstituciones.csv")#encoding="latin-1")
df
df.year=df.year.astype(int)
df
Out[179]:
In [180]:
df.columns
Out[180]:
In [181]:
dfi = pd.read_csv("Cluster.cluster.csv") #encoding="latin-1")
dfi
Out[181]:
In [182]:
dfi.columns
Out[182]:
In [183]:
clusters = dfi["Cluster"]
In [184]:
dfm = df.join(clusters) #append 'Cluster' to 'df'
dfm
Out[184]:
In [185]:
dfm.year=dfm.year.astype(int)
dfm
Out[185]:
In [186]:
dfm.columns
Out[186]:
In [187]:
pd.plotting.bootstrap_plot(dfm['year'])
Out[187]:
Cluster Description¶
In [188]:
fig = px.scatter(dfm, x="year", y="articulos", color="Cluster", facet_col="Cluster")
fig.show()
In [189]:
fig4 = px.box(dfm,y='year',x='Cluster', color='Cluster')
fig4
In [190]:
fig4 = px.scatter(dfm,y='year',x='Cluster', color='Instituciones')
fig4
In [191]:
fig4 = px.scatter(dfm,x='year',y='Instituciones', color='Cluster')
fig4
In [192]:
# all(round(dfm.year) != df.year)
## g = round(dfm.year) != dfm.year
## g.describe()
### g.value_counts('True')
Cluster 0¶
In [193]:
df0 = dfm[dfm.Cluster == 0]
df0
Out[193]:
In [194]:
fig = px.box(df0,x='year', y=['articulos','capitulos','trabajos_grado','innovaciones', 'libros', 'softwares'], points="all", color="Cluster")
fig.show()
In [195]:
fig = px.box(df0,x='year', y=['articulos','capitulos','trabajos_grado','innovaciones', 'libros', 'softwares'], points="all", color="Instituciones")
fig.show()
In [196]:
fig = px.scatter(df0, x="year", y="articulos", color="Instituciones", size="articulos")
fig.show()
Cluster 1¶
In [197]:
df1 = dfm[dfm.Cluster == 1]
df1
Out[197]:
In [198]:
fig = px.box(df1,x='year', y=['articulos','capitulos','trabajos_grado','innovaciones', 'libros', 'softwares'], points="all", color="Cluster")
fig.show()
In [199]:
fig = px.box(df1,x='year', y=['articulos','capitulos','trabajos_grado','innovaciones', 'libros', 'softwares'], points="all", color="Instituciones")
fig.show()
In [200]:
fig = px.scatter(df1, x="year", y="articulos", color="Instituciones", size="articulos")
fig.show()
Cluster 2¶
In [201]:
df2 = dfm[dfm.Cluster == 2]
df2
Out[201]:
In [202]:
fig = px.box(df2,x='year', y=['articulos','capitulos','trabajos_grado','innovaciones', 'libros', 'softwares'], points="all", color="Cluster")
fig.show()
In [203]:
fig = px.box(df2,x='year', y=['articulos','capitulos','trabajos_grado','innovaciones', 'libros', 'softwares'], points="all", color="Instituciones")
fig.show()
In [204]:
fig = px.scatter(df2, x="year", y="articulos", color="Instituciones", size="articulos")
fig.show()
Cluster 3¶
In [205]:
df3 = dfm[dfm.Cluster == 3]
df3
Out[205]:
In [206]:
fig = px.box(df3,x='year', y=['articulos','capitulos','trabajos_grado','innovaciones', 'libros', 'softwares'], points="all", color="Cluster")
fig.show()
In [207]:
fig = px.box(df3,x='year', y=['articulos','capitulos','trabajos_grado','innovaciones', 'libros', 'softwares'], points="all", color="Instituciones")
fig.show()
In [208]:
fig = px.scatter(df3, x="year", y="articulos", color="Instituciones", size="articulos")
fig.show()
Cluster 4¶
In [209]:
df4 = dfm[dfm.Cluster == 4]
df4
Out[209]:
In [210]:
fig = px.box(df4,x='year', y=['articulos','capitulos','trabajos_grado','innovaciones', 'libros', 'softwares'], points="all", color="Cluster")
fig.show()
In [211]:
fig = px.box(df4,x='year', y=['articulos','capitulos','trabajos_grado','innovaciones', 'libros', 'softwares'], points="all", color="Instituciones")
fig.show()
In [212]:
fig = px.scatter(df4, x="year", y="articulos", color="Instituciones", size="articulos")
fig.show()
Producción¶
In [213]:
dfy = dfm.year.value_counts(ascending=False).reset_index()
dfgy=dfy.round({"index":0, "year":2})
dfgy.columns =['Year', 'Cantidad']
In [214]:
dfgy.Year.value_counts(ascending=False).reset_index().unique=True
dfgy.sort_values('Year')
fig4 = px.scatter(dfgy,y ='Cantidad',x='Year')
fig4
In [215]:
dfgy = dfgy.sort_values(by=['Year'], ascending=False)
In [216]:
fgy = sns.pairplot(dfgy)
In [217]:
dfgy.columns
Out[217]:
In [218]:
# Cantidad años
conteo_años = dfgy.groupby('Year').size()
conteo_años.values
Out[218]:
In [219]:
años = dfgy.groupby('Year').sum().reset_index()[['Year','Cantidad']]
años
fig = px.bar(años, x="Year", y ='Cantidad', title="Años",text_auto=True)
fig.show()
In [220]:
fig4 = px.scatter(años,y ='Cantidad',x='Year', marginal_x='histogram',marginal_y='violin', trendline='ols')
fig4
In [221]:
años['log_cant'] = np.log(años.Cantidad)
fig4 = px.scatter(años,y ='log_cant',x='Year', marginal_x='histogram',marginal_y='violin', trendline='ols')
fig4
In [222]:
fig = px.scatter(años, x="Year", y="log_cant", size="log_cant")
fig.show()
In [223]:
yy = sns.pairplot(años)
In [224]:
fig = px.line(años, x="Year", y="log_cant")
fig.show()
In [225]:
fig = px.scatter(años, x="Year", y="Cantidad", marginal_y="box")
fig.show()
In [226]:
años
Out[226]:
In [227]:
años.index
Out[227]:
In [228]:
np.log(años.index)
Out[228]:
In [229]:
fig4 = px.line(años,y ='log_cant',x='Year')
fig4
Grupos de investigacion¶
Data¶
Cluster.infoGrupos.csv https://drive.google.com/drive/u/0/folders/1APVnfD1zPgTHFAr0YMG6UMTwZ52MQVa9
In [230]:
dfg = pd.read_csv("Cluster.infoGrupos.csv")
dfg
Out[230]:
In [231]:
dfg.columns
Out[231]:
In [232]:
dfg.T
Out[232]:
In [233]:
dfg ['year'].plot()
Out[233]:
In [234]:
pd.plotting.bootstrap_plot(dfg['year'])
Out[234]:
In [235]:
dfgy1 = dfg.year.value_counts(ascending=False).reset_index()
#dfgy1=dfgy1.round({"index":0, "year":2})
dfgy1.columns =['Year', 'Cantidad']
In [236]:
dfgy1.Year.value_counts(ascending=False).reset_index().unique=True
dfgy1.sort_values('Year')
fig4 = px.scatter(dfgy1,y ='Cantidad',x='Year')
fig4
In [237]:
años1 = dfgy1.groupby('Year').sum().reset_index()[['Year','Cantidad']]
años1
fig = px.bar(años1, x="Year", y ='Cantidad', title="Años",text_auto=True)
fig.show()
In [238]:
años1['log_cant'] = np.log(años1.Cantidad)
fig4 = px.scatter(años1,y ='log_cant',x='Year', marginal_x='histogram',marginal_y='violin', trendline='ols')
fig4
Articulos¶
In [239]:
#articulos = pd.read_csv('/Users/jmlz_rp/Documents/BIOS/BIOS_rar/Datos/Datos sin procesar/Categorias minciencias/articulos.csv')
articulos = pd.read_csv('articulos.csv')
articulos
Out[239]:
In [240]:
articulos.columns
Out[240]:
In [241]:
a = articulos.anio.sort_values(ascending=True)
In [242]:
a.tail()
Out[242]:
In [243]:
a.head()
Out[243]:
In [244]:
a.describe()
Out[244]:
In [245]:
a.shape
Out[245]:
In [246]:
fig = px.scatter(articulos,x='anio', y='grupo')
fig.update_xaxes(range=[1900,2025])
fig